Full-term pregnancy early in life reduces the lifetime risk of breast cancer in women, and in rodents protects against chemically induced mammary cancer, suggesting that rodents may serve as a good model for this effect in humans. However, the effect of pregnancy on transgene-induced carcinogenesis cannot yet be determined in transgenic mice. Traditional transgenic mouse models of mammary cancer employ gene-targeting elements that are upregulated by hormones at puberty, during pregnancy and lactation; the upregulation of transgene expression by pregnancy complicate attempts to study the effects of pregnancy on transgene-induced mammary cancer. In addition, transgene expression in traditional transgenic mice occurs in nearly all mammary epithelial cells, and is active throughout life. This feature does not mimic the spatial pattern of carcinogenic gene alterations in human breast, which typically is focal rather than widespread. Our purpose is to develop a novel transgenic approach that can be used to precisely and independently regulate when transgene expression is "turned on" in the mammary epithelium (specifically, after the completion of pregnancy), and in a focal pattern that mimics the upregulation of breast cancer relevant genes in human disease. To accomplish these goals, we will establish transgenic mice with following characteristics: (1) timing of onset of TGFa expression is regulatable, (2) TGFa expression is activated irreversibly in a controllable fraction of mammary epithelial cells, and (3) TGFa is under the control of an estrogen and prolactin nonresponsive promoter that permits evaluation of the effect on lesion development of hormonal modulation. This application has the following specific aims: Aim 1. Develop transgenic mice that permit manipulation of TGFa expression in mammary epithelium. Aim 2. Establish how prior pregnancy affects TGFa -induced mammary tumorigenesis.